CN105913429B - The calculation method of intelligent terminal user visual perception response time delay index - Google Patents
The calculation method of intelligent terminal user visual perception response time delay index Download PDFInfo
- Publication number
- CN105913429B CN105913429B CN201610223448.9A CN201610223448A CN105913429B CN 105913429 B CN105913429 B CN 105913429B CN 201610223448 A CN201610223448 A CN 201610223448A CN 105913429 B CN105913429 B CN 105913429B
- Authority
- CN
- China
- Prior art keywords
- input image
- image sequence
- sequence
- variance
- picture numbers
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
Abstract
The present invention provides a kind of calculation method of intelligent terminal user visual perception response time delay index, comprising: original sequence is carried out gray processing processing, generates input image sequence;Target detection is carried out in input image sequence, and the coordinate of target is obtained in the multiple input image sequences for detecting target;The extreme value in each coordinate is calculated, the serial number of the corresponding input picture of extreme value is recorded as the first picture numbers;The first subsequence of multiple groups is filtered out from input image sequence according to the first adaptive threshold, and by being ranked up to the first subsequence of multiple groups, determines third picture numbers;The second subsequence of multiple groups is filtered out from input image sequence according to the second adaptive threshold, and determines the second picture numbers from the second subsequence of multiple groups according to the first picture numbers, third picture numbers;According to the first picture numbers and the second picture numbers computing terminal response time;The application response time is calculated according to third picture numbers and the second picture numbers.
Description
Technical field
The present invention relates to mobile terminals to evaluate and test technology, specifically ringing about a kind of intelligent terminal user visual perception
Answer the calculation method of time delay index.
Background technique
The new change of intelligent terminal especially smart phone requires adaptable therewith intelligent terminal evaluation and test system and evaluation and test
Method, it is necessary to jump out the test of traditional concern communication capacity, study the test method of service-oriented carrier and service platform.
User experience is core index of comprehensive, effective evaluation and test smart terminal product as business carrier and service platform true value,
Carry out the test based on user experience and obtains the generally approval of industry.
Evaluation and test system comprehensive not yet domestic at present and evaluation and test ability, carry out the intelligent terminal evaluation and test based on user experience
Research forms intelligent terminal user experience evaluation and test system and evaluation and test ability, can establish user experience evaluating standard for industry and mention
It for Test Suggestion, can also be used in intelligent terminal entry test, industry authentication test, manufacturer is promoted actively to promote user's body
It tests.
The whole impressions of user experience, i.e. user before using a product or system, during use and after use,
Including the various aspects such as emotion, faith, hobby, cognition impression, physiology and psychoreaction, behavior and achievement.Three influence users
The factor of experience are as follows: system, user and use environment.Therefore, user experience can be defined as to user and product or service interaction
Whole subjective feelings caused by the process.
In intelligent terminal user experience, especially visual perception field, the test method of current main-stream are exactly user's ginseng
With assessment, this index of speed is perceived for user, without the quantitative to describe of certain objectivity, this results in test to tie
The unstability and nonrepeatability of fruit, and the sensibility of the conditions such as terminal type is tested for tested crowd.
The method evaluated and tested in the prior art to user experience mainly includes two kinds.One of which is that embedded system is rung
Performance evaluation is answered, i.e., is obtained by way of reading log recording and outputs and inputs the time difference, obtain relevant delay performance, is led to
Installation test software is crossed, frame number when testing intelligent terminal system interface or application software operation in the unit time obtains smoothness
Performance, that is, response delay index.This test method reads log, and time-consuming, and third party software is needed to access;Response delay essence
It spends low, has no idea to distinguish intelligent terminal response time and application response time.
Another kind is that measured terminal is fixed on to the station for being equipped with high-speed camera, is opened using manipulator and applies journey
Sequence starts simultaneously at timing.The picture of shooting is reached server and carries out image comparison by high-speed camera, is matched to application program and is rung
Stop timing after the normal pictures of the page should be finished, calculates time difference, that is, application response time.It is applied in this method
Matching template need to shift to an earlier date Manual interception, and matching threshold is not fixed, and have no idea to distinguish intelligent terminal response time and application program
Response time can not continuously repeat test.
Summary of the invention
The main purpose of the embodiment of the present invention is to provide a kind of intelligent terminal user visual perception response time delay index
Calculation method, to solve problems of the prior art, objective computation user's visual perception response time delay index improves response
The computational accuracy of time delay index.
To achieve the goals above, the embodiment of the present invention provides a kind of intelligent terminal user visual perception response time delay index
Calculation method, the calculation method include: by original sequence carry out gray processing processing, generate input image sequence;
Target detection is carried out in the input image sequence, and the target is obtained in the multiple input image sequences for detecting target
Coordinate;The extreme value in each coordinate is calculated, the serial number of the corresponding input picture of the extreme value is recorded as the first image sequence
Number;The first subsequence of multiple groups is filtered out from the input image sequence according to the first adaptive threshold, and to the multiple groups first
Subsequence is ranked up, to determine third picture numbers;It is filtered out according to the second adaptive threshold from the input image sequence
The second subsequence of multiple groups, and determined from second subsequence of multiple groups according to the first image serial number, third picture numbers
Second picture numbers;According to the first image serial number and the second picture numbers computing terminal response time;According to the third
Picture numbers and the second picture numbers calculate the application response time.
In one embodiment, above-mentioned that original sequence is subjected to gray processing processing, input image sequence is generated, specifically
Include: that gray processing processing is carried out to the original sequence, generates the grey level histogram of the original sequence;According to institute
It states grey level histogram and judges gray processing treated whether the contrast of original sequence is less than or equal to the first preset threshold;
If it is, to the gray processing, treated that original sequence is normalized, and generates the input picture sequence
Column;Otherwise, using the gray processing treated original sequence as the input image sequence.
In one embodiment, by following formula, to the gray processing, treated that place is normalized in original sequence
Reason:
Wherein, x is the abscissa of the i-th frame input image sequence;Y is the ordinate of the i-th frame input image sequence; Ii(x,
It y) is the gray value of the i-th frame input image sequence;IminFor the gray value of pixel in the i-th frame input image sequence
Minimum value;ImaxFor the maximum value of the gray value of pixel in the i-th frame input image sequence;Step is drawing coefficient, 0 <
Step < 255.
In one embodiment, above-mentioned that target detection is carried out in the input image sequence, it specifically includes: utilizing template
Matching process carries out mechanical arm target detection to the input image sequence, obtains the multiple input picture sequences for detecting target
Column;Or finger target detection is carried out to the input image sequence using Otsu threshold method, acquisition detects the multiple of target
Input image sequence.
In one embodiment, the step of determining first adaptive threshold and the second adaptive threshold includes: according to
I frame input image sequence and i+1 frame input image sequence calculating difference image sequence, and further calculate the error image
The variance of sequence generates variance sequence;The variance of one group of preset quantity is chosen from the stem of the variance sequence, forms first group
Variance, and from the variance of preset quantity described in one group of the selection of the tail portion of the variance sequence, form second group of variance;By described
Maximum value in one group of variance is set as the first word adaptive threshold, and the minimum value in first group of variance is set as second
Described first sub- adaptive threshold and the second sub- adaptive threshold are determined as first adaptive thresholding by sub- adaptive threshold
Value;Maximum value in second group of variance is set as the sub- adaptive threshold of third, by the minimum in second group of variance
Value is set as the 4th sub- adaptive threshold, and the sub- adaptive threshold of the third and the 4th sub- adaptive threshold are determined as described
Two adaptive thresholds.
In one embodiment, above-mentioned that multiple groups first are filtered out from the input image sequence according to the first adaptive threshold
Subsequence, and first subsequence of multiple groups is ranked up, it to determine third picture numbers, specifically includes: by the variance
It is greater than the 4th sub- adaptive threshold in sequence and is less than the serial number composition of the multiple groups variance of the sub- adaptive threshold of the third
First position array positionend;To the first position array positionendDiff is carried out, the first difference is generated
As a result Diffend(j), wherein Diffend(j)=positionend(j+1)-positionend(j), j is the variance sequence
Position;By the first difference result Diffend(j) the corresponding position of value in greater than one second preset threshold is charged to after adding 1
Brush screen judges array Pos;When the brush screen judges array Pos for sky, by the first difference result Diffend(j) it is less than in
The serial number of third predetermined threshold value charges to the first tail portion array pend;By the first tail portion array pendIn minimum value corresponding to
The serial number of input picture is determined as the third picture numbers.
In one embodiment, it when it is sky that the brush screen, which judges array Pos not, obtains the brush screen and judges in array Pos
The first maximum value frame3start;By the first difference result DiffendIn be less than the third predetermined threshold value serial number remember
Enter the second tail portion array p 'end;By the second tail portion array p 'endIn be greater than or equal to the first maximum value frame3start
The serial number of corresponding serial number charges to third tail portion array pend1;By the third tail portion array pend1In minimum value be determined as institute
State third picture numbers.
In one embodiment, it is calculated by the following formula the error image sequence:
Di(x, y)=Ii+1(x, y)-Ii(x, y),
Wherein, x is the abscissa of the input image sequence;Y is the ordinate of the input image sequence;Ii(x, y)
For the gray value of the i-th frame input image sequence;Ii+1(x, y) is the gray value of the i+1 frame input image sequence;With
And it is calculated by the following formula the variance of the error image sequence:
Wherein, n=h*w, h and w are respectively the height and width of the input image sequence,
In one embodiment, above-mentioned that multiple groups second are filtered out from the input image sequence according to the second adaptive threshold
Subsequence, and the second image is determined from second subsequence of multiple groups according to the first image serial number, third picture numbers
Serial number specifically includes: median filtering carried out to the variance sequence by an adaptive template, generates filtering rear difference sequence,
The size of the adaptive template are as follows: template=(frame3-frame1) %10, wherein frame1For the first image
Serial number, frame3For the third picture numbers;It obtains and is greater than the described second sub- adaptive thresholding in the difference sequence of the filtering rear
The filtering variance serial number of value and the multiple groups variance less than the described first sub- adaptive threshold, and further by the filtering variance
It is greater than the first image serial number in serial number and is less than the serial number composition second position array of the third picture numbers
positionstart;To the second position array positionstartDiff is carried out, the second difference result is generated
Diffstart(k), wherein Diffstart(k)=positionstart(k+1)-positionstart(k), k is the variance sequence
Position;By the second difference result Diffstart(k) serial number in less than the third predetermined threshold value charges to stem array
pstart;By the stem array pstartIn maximum value corresponding to the serial number of input picture be determined as second image
Serial number.
In one embodiment, above-mentioned when being responded according to the first image serial number and the second picture numbers computing terminal
Between, specifically include: the terminal response time, is equal to second picture numbers corresponding time and the first image serial number
The difference of corresponding time.
In one embodiment, above-mentioned to calculate application response according to the third picture numbers and the second picture numbers
Time specifically includes: the application response time is equal to the third picture numbers corresponding time and second figure
As the difference of serial number corresponding time.
The beneficial effect of the embodiment of the present invention is, the response delay index in user's visual perception is objectified and measured
Change, calculated results are more accurate, and can repeat to evaluate and test by this method, develop and evaluate and test for intelligent terminal evaluating system
Platform construction provides valuable thinking and suggestion.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, embodiment will be described below
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is the process of the calculation method of the intelligent terminal user visual perception response time delay index of the embodiment of the present invention
Figure;
Fig. 2 is that the normalized of the embodiment of the present invention enhances contrast results schematic diagram;
Fig. 3 is the object detection results schematic diagram of the embodiment of the present invention;
Fig. 4 is the conversion process schematic diagram of the corresponding image of the first picture numbers of acquisition of the embodiment of the present invention;
Fig. 5 is the variance change curve of the input image sequence of the embodiment of the present invention;
Fig. 6 A to Fig. 6 I is the conversion process schematic diagram of the corresponding image of acquisition third picture numbers of the embodiment of the present invention;
Fig. 7 is the median-filtered result schematic diagram of the embodiment of the present invention;
Fig. 8 is the conversion process schematic diagram of the corresponding image of the second picture numbers of acquisition of the embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The embodiment of the present invention provides a kind of calculation method of intelligent terminal user visual perception response time delay index.It ties below
Closing attached drawing, the present invention is described in detail.
The embodiment of the present invention provides a kind of calculation method of intelligent terminal user visual perception response time delay index, such as Fig. 1
Shown, which mainly includes following steps:
Step S101: original sequence is subjected to gray processing processing, generates input image sequence;
Step S102: carrying out target detection in input image sequence, in the multiple input image sequences for detecting target
The middle coordinate for obtaining target;
Step S103: calculating the extreme value in each coordinate, and the corresponding input image sequence of extreme value is recorded as the first image sequence
Number;
Step S104: the first subsequence of multiple groups is filtered out from input image sequence according to the first adaptive threshold, and is passed through
The first subsequence of multiple groups is ranked up, determines third picture numbers;
Step S105: filtering out the second subsequence of multiple groups from input image sequence according to the second adaptive threshold, and according to
First picture numbers, third picture numbers determine the second picture numbers from the second subsequence of multiple groups;
Step S106: according to the first picture numbers and the second picture numbers computing terminal response time;
Step S107: the application response time is calculated according to third picture numbers and the second picture numbers.
S101 to step S107 through the above steps, the intelligent terminal user visual perception response time delay of the embodiment of the present invention
The calculation method of index pre-processes the image sequence of input by gray processing, obtains target by target detection method
Position simultaneously judges the application triggers moment.By solving variance curve to pretreatment image sequence and judging that variance curve is steady
Section, so that computational intelligence terminal begins to respond to the moment.Judge that the moment is completed in response using adaptive threshold method.Response delay refers to
The response time of intelligent terminal in mark is that intelligent terminal begins to respond to the moment and the application triggers moment makes the difference result;Using
The program response time is that intelligent terminal begins to respond to the moment and the response completion moment makes the difference result.As it can be seen that the embodiment of the present invention
The calculation method of intelligent terminal user visual perception response time delay index, by user's visual perception response delay index (
In the embodiment of the present invention, i.e., the response time and application response time of above-mentioned intelligent terminal) it objectifies and quantifies, gained
Calculated result is more accurate, and can repeat to evaluate and test by this method, and platform is developed and evaluated and tested for intelligent terminal evaluating system and is built
If providing valuable thinking and suggestion.
Below in conjunction with attached drawing to the calculating side of the intelligent terminal user visual perception response time delay index of the embodiment of the present invention
Each step of method is described in detail.
Original sequence is carried out gray processing processing, generates input image sequence by above-mentioned step S101.
The N frame original sequence that high speed acquisition is obtained carries out gray processing processing, IiIt indicates after gray processing is handled
The i-th frame image.Statistical picture IiGrey level histogram, when the range (i.e. the contrast of image) of greyscale transformation is less than or equal to
(first preset threshold is to judge gray value maximum value and gray value minimum value in the i-th frame image to one first preset threshold
Between otherness first preset threshold can be set as 50, but the present invention is not limited thereto in this embodiment) when,
Image I is stretched with common scale using after normalized function normalizediGreyscale transformation range enhance contrast.Contrast
Normalize stretch function are as follows:Wherein, x is the abscissa of the i-th frame input image sequence;y
For the ordinate of the i-th frame input image sequence;Ii(x, y) is the gray value of the i-th frame input image sequence;IminIt is described
The minimum value of the gray value of pixel in i-th frame input image sequence;ImaxFor the ash of pixel in the i-th frame input image sequence
The maximum value of angle value;Step is drawing coefficient, 0 < step < 255, and the specific value of step can be modified according to actual needs, one
As be the larger value (such as 200) close to 255.Using the image sequence after normalized as input image sequence.On
The result of normalized is stated as shown in Fig. 2, wherein Fig. 2 (a) show the image before normalized, Fig. 2 (b) show through
Image after crossing normalized.
It, then can directly will be by gray processing treated image when the range of greyscale transformation is greater than first preset threshold
Sequence is as input image sequence.
Above-mentioned steps S102 carries out target detection in input image sequence, in the multiple input pictures for detecting target
The coordinate of target is obtained in sequence.
Specifically carrying out mechanical arm target detection to input image sequence using template matching method, acquisition is detected
Multiple input image sequences of target;Or finger target detection is carried out to input image sequence using Otsu threshold method, it obtains
Take the multiple input image sequences for detecting target.As shown in Figure 3, wherein Fig. 3 (a) show mapping to be checked, shown in Fig. 3 (b)
It is the image comprising target (such as finger) shown in Fig. 3 (b) for object detection results schematic diagram.Further,
Obtain the coordinate of target in the multiple input image sequences for detecting target.
After obtaining the coordinate of target, S103, calculates the extreme value in each coordinates of targets through the above steps, finds out the pole
It is worth the serial number of corresponding image, the corresponding image of the serial number is that application program is triggered moment image, and application program is touched
Hair moment image is recorded as the first picture numbers, which is denoted as frame1.As shown in Figure 4, wherein Fig. 4 (a)
It show finger and shifts to image during application program (call), when Fig. 4 (b) show finger down application program (call)
It carves (the corresponding coordinate in the position of the finger detected at this time using Otsu threshold method is above-mentioned extreme value), shown in Fig. 4 (c)
When just having left application program (call) for finger, Fig. 4 (d) show finger and leaves figure during application program (call)
Picture;Fig. 4 (e) show mechanical arm and shifts to image during application program (call), and Fig. 4 (f), which show mechanical arm and presses, to be answered
With program (call) moment, (the corresponding coordinate in the position of the mechanical arm detected at this time using template matching method is above-mentioned
Extreme value), when Fig. 4 (g) show mechanical arm just and to leave application program (call), Fig. 4 (h) show mechanical arm and leaves using journey
Image during sequence (call).
Above-mentioned steps S104 filters out the first subsequence of multiple groups from input image sequence according to the first adaptive threshold, and
By being ranked up to the first subsequence of multiple groups, third picture numbers are determined.
For above-mentioned input image sequence, the variance sequence Var of the input image sequence is sought first.
Specifically being calculated by the following formula the error image sequence:
Di(x, y)=Ii+1(x, y)-Ii(x, y),
Wherein, x is the abscissa of input image sequence;Y is the ordinate of input image sequence;Ii(x, y) is the i-th frame
The gray value of input image sequence;Ii+1(x, y) is the gray value of i+1 frame input image sequence.
Then, it is calculated by the following formula the variance of above-mentioned error image sequence:
Wherein, n=h*w, h and w are respectively the height and width of input image sequence,The variance
Change curve can respectively correspond surface chart when two kinds of different intelligent terminal triggering application programs as shown in Fig. 5 (a) and Fig. 5 (b)
The variance of picture changes, wherein Fig. 5 (a) show that there is no the surface charts that the intelligent terminal of brush screen situation triggering application program is
The variance change curve of picture, Fig. 5 (b) show that there are the intelligent terminal of the brush screen situation triggering application program interface images that are
Variance change curve.
Specifically, adaptive threshold (including the first adaptive threshold and the second adaptive thresholding are obtained by following procedure
Value): the variance of one group of preset quantity is chosen from the stem of above-mentioned variance sequence, forms first group of variance, and from variance sequence
Tail portion choose one group of preset quantity variance, form second group of variance;Maximum value in first group of variance is set as first
Minimum value in first group of variance is set as the second sub- adaptive threshold by sub- adaptive threshold, by the first sub- adaptive threshold
And second sub- adaptive threshold be determined as the first adaptive threshold;It is adaptive that maximum value in second group of variance is set as third
Threshold value is answered, the minimum value in second group of variance is set as the 4th sub- adaptive threshold, by the sub- adaptive threshold of third and the 4th
Sub- adaptive threshold is determined as the second adaptive threshold.
For example, in one embodiment, above-mentioned preset quantity is 15, then in this embodiment, variance sequence Var is choseni
15 groups of variances of stem form first group of above-mentioned variance, choose variance sequence VariTail portion 15 groups of variances composition it is above-mentioned
Second group of variance.It is corresponding, the sub- adaptive threshold start of above-mentioned firstmax=max { Var1, Var2, Var3...,
Var15, the sub- adaptive threshold start of above-mentioned secondmin=max { Var1, Var2, Var3..., Var15, above-mentioned third
Adaptive threshold endmax=max { VarN-15, VarN-14, VarN-13..., VarN-1, the 4th above-mentioned sub- adaptive threshold
endmin=min { VarN-15, VarN-14, VarN-13..., VarN-1}。
It should be noted that be by 15 in above content as preset quantity, but in practical applications, it can be according to practical need
The numerical value of the preset quantity is adjusted, the present invention is not limited thereto.
Screen shows two states after application program reaches and terminates responsive state, a kind of such as Fig. 6 A, Fig. 6 B and Fig. 6 C
It is shown, brush screen situation is not present, wherein Fig. 6 A and Fig. 6 B show the interface in application program (call) response, is shown in Fig. 6 C
Interface, which just responds, completes the moment;One kind is as shown in Fig. 6 D to Fig. 6 I, and there are brush screen situations, and wherein Fig. 6 D and Fig. 6 E show and answer
It is that interface just responds the completion moment shown in Fig. 6 F with the interface in program (call) response;Fig. 6 G, Fig. 6 H and Fig. 6 I are shown
Brush screen situation existing for intelligent terminal interface after the completion of application response.
For variance sequence Var, array positionendIn record all Var (j) > endminAnd Var (j) < endmax
Subscript j, array positionendIn value be response complete moment possible position.Remember DiffendIt (j) is pair
positionendSolve diff as a result, record Diffend(j) be greater than one second preset threshold (e.g. 10, can basis
Actual needs is adjusted, and the present invention is not limited thereto) position, and array Pos is stored in after the position is added 1, if array
Pos is sky, then there are above-mentioned brush screen situations for explanation, at this point, finding Diffend(j)≤2 point, corresponding subscript is charged to
Array pend, minimum value min (p in all satisfactory subscriptsend) corresponding image is that application program is just completed
The position of response, i.e. third picture numbers, third picture numbers frame3=min (pend).If array Pos is not sky,
Illustrate that there is no above-mentioned brush screen situations, at this point, seeking the possible initial position frame of third picture numbers first3start,
Middle frame3start=max (Pos) finds Diffend(j)≤2 corresponding serial number is charged to array p by pointend, in array
pend1Middle deposit is all to be greater than or equal to frame3startThe value of corresponding serial number, third picture numbers are all satisfactory
Minimum value min (p in positionend1) corresponding image, it is the position that application program just completes response, the third image
Serial number frame3=min (pend1)。
For the content shown in above-mentioned Fig. 5, wherein Fig. 5 (a) show that there is no the triggerings of the intelligent terminal of brush screen situation to answer
With the variance change curve for the interface image that program is.It may at this point, seeking third picture numbers by adaptive threshold first
Initial position frame3start, as shown in Fig. 5 (a), the data that above-mentioned array Pos is included should be serial number 100-110,
In 150-160,250-260 (by taking each steady section includes 10 variances as an example) difference result greater than 10 as a result, serial number 110
The corresponding position in serial number 100-110,150-160,250-260 of variance is the 10th, and the variance of serial number 160 is corresponding
Position in serial number 100-110,150-160,250-260 is the 20th, therefore, Diffend(10)=positionend
(11)-positionend(10)=150-110=40, Diffend(20)=positionend(21)-positionend(20)=
250-160=90, it is seen then that Diffend(10) and Diffend(20) it is all larger than second preset threshold (10), at this point, by position
10,20 plus 1 after be stored in array Pos, then the data in array Pos be 11,21.At this point, frame3start=max (Pos)=21.
Then, Diff is foundend(j)≤2 corresponding serial number is charged to array p by pointend, in array pend1Middle deposit is all to be greater than
Or it is equal to frame3startThe value of corresponding serial number, wherein frame3startCorresponding serial number is serial number corresponding to 21, that is, is schemed
250 shown in 5 (a).Third picture numbers are the minimum value min (p in all satisfactory positionsend1) corresponding figure
Picture is the position that application program just completes response, third picture numbers frame3=min (pend1)。
Fig. 5 (b) show the variance variation for the interface image for being there are the intelligent terminal of brush screen situation triggering application program
Curve.At this point, first looking for Diffend(j)≤2 corresponding subscript is charged to array p by pointend, false as shown in Fig. 5 (b)
If the starting point of subsequent steady section is serial number 160, then the subscript being stored in the array is 160-300.Then, all to meet
State the minimum value min (p in the subscript of requirementend) (namely 160) corresponding image is that application program is just completed to respond
Position, i.e. third picture numbers, third picture numbers frame3=min (pend)。
Above-mentioned step S105 filters out the second subsequence of multiple groups from input image sequence according to the second adaptive threshold,
And the second picture numbers are determined from the second subsequence of multiple groups according to the first picture numbers, third picture numbers.
Other side's difference sequence Var first carries out adaptive template size median filtering and removes removal of impurities spot noise, the adaptive template
Size template=(frame3-frame1) %10, result is Varmed after note filtering, and median filtering removes removal of impurities spot noise
As a result as shown in fig. 7, wherein Fig. 7 (a) show original-party difference sequence waveform diagram, Fig. 7 (b) is shown to be denoised by median filtering
Waveform diagram afterwards.Array positionstartIn record all Varmed (k) > startminAnd Varmed (k) < startmax's
Subscript k, also, k meets frame1< k < frame3, positionstartIn value be begin to respond to moment possible position.
Remember DiffstartFor to positionstartSolve diff as a result, find Diffstart(j)≤2 point, will be corresponding
Serial number charges to array pstart, maximum value max (p in all satisfactory positionsstart) the i.e. corresponding intelligence of corresponding image
Terminal just completes response position, i.e. the second picture numbers, the second picture numbers frame2=max (pstart).Intelligent terminal
The situation of change for completing the interface image of response process is as shown in Figure 8, wherein Fig. 8 (a), Fig. 8 (b) and Fig. 8 (c) are shown not
There are the variation that the intelligent terminal of brush screen situation completes the interface image of response process, Fig. 8 (d), Fig. 8 (e) and Fig. 8 (f) are shown
For there are the variation of the interface image of the intelligent terminal of brush screen situation completion response process, (the light and shade striped that brush screen generates is not stopping
Ground is mobile).Fig. 8 (a) and Fig. 8 (d), which show terminal, will begin to respond to interface, and it is rigid that Fig. 8 (b) and Fig. 8 (e) show terminal
Rigid response interface, Fig. 8 (c) and Fig. 8 (f) show the figure during terminal response application program process median surface continuous transformation
Picture.
After obtaining the first picture numbers, the second picture numbers and third picture numbers by above steps
Calculate response delay index.
It should be noted that being by seeking between each frame of input image sequence in above-mentioned steps S104 and step S105
Variance mode come obtain image sequence transformation severe degree.But in practical applications, can also by other calculations come
Indicate that image sequence converts severe degree, such as derivation at each inflection point for passing through the gray-value variation curve to image sequence,
The situation of change of gray value is obtained by way of derivation, so that it is violent to obtain each frame image sequence transformation of input image sequence
Degree, or each frame image sequence transformation severe degree of input image sequence is obtained by other calculations, the present invention is simultaneously
It is not limited.
Above-mentioned steps S106, according to the first picture numbers and the second picture numbers computing terminal response time.Specifically, should
Terminal response time, is equal to the difference of time corresponding with the first picture numbers the second picture numbers corresponding time.
Above-mentioned steps S107 calculates the application response time according to third picture numbers and the second picture numbers.Specifically
Ground, the application response time are equal to the difference of time corresponding with the second picture numbers third picture numbers corresponding time
Value.
The present invention is using the acquisition of high speed acquisition camera and the image sequence during intelligent terminal interactive, by scheming to acquisition
Picture sequence analysis detection intelligent terminal state, will using the calculating of the stronger target detection of simple and universality and variance index
The calculation method automation of response delay index in the experience of intelligent terminal user visual perception.The present invention is intelligent terminal user body
The practical applications such as formulation, the test of the objective performance indicator tested provide effective tool, have a vast market foreground and apply
Value.
Those of ordinary skill in the art will appreciate that implementing the method for the above embodiments can lead to
Program is crossed to instruct relevant hardware and complete, which can be stored in a computer readable storage medium, such as
ROM/RAM, magnetic disk, CD etc..
Particular embodiments described above has carried out further in detail the purpose of the present invention, technical scheme and beneficial effects
Describe in detail it is bright, it should be understood that the above is only a specific embodiment of the present invention, the guarantor being not intended to limit the present invention
Range is protected, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in this
Within the protection scope of invention.
Claims (11)
1. a kind of calculation method of intelligent terminal user visual perception response time delay index, which is characterized in that the calculating side
Method includes:
Original sequence is subjected to gray processing processing, generates input image sequence;
Target detection is carried out in the input image sequence, in the multiple input image sequences for detecting target described in acquisition
The coordinate of target;
The extreme value in each coordinate is calculated, the serial number of the corresponding input picture of the extreme value is recorded as the first picture numbers;
The first subsequence of multiple groups is filtered out from the input image sequence according to the first adaptive threshold, and to the multiple groups first
Subsequence is ranked up, to determine third picture numbers;
The second subsequence of multiple groups is filtered out from the input image sequence according to the second adaptive threshold, and according to first figure
As serial number, third picture numbers determine the second picture numbers from second subsequence of multiple groups;
According to the first image serial number and the second picture numbers computing terminal response time;
The application response time is calculated according to the third picture numbers and the second picture numbers.
2. the calculation method of intelligent terminal user visual perception response time delay index according to claim 1, feature exist
In by original sequence progress gray processing processing, generation input image sequence is specifically included:
Gray processing processing is carried out to the original sequence, generates the grey level histogram of the original sequence;
Gray processing is judged treated whether the contrast of original sequence is less than or equal to one according to the grey level histogram
First preset threshold;
If it is, to the gray processing, treated that original sequence is normalized, and generates the input figure
As sequence;Otherwise, using the gray processing treated original sequence as the input image sequence.
3. the calculation method of intelligent terminal user visual perception response time delay index according to claim 2, feature exist
In by following formula, to the gray processing, treated that original sequence is normalized:
Wherein, x is the abscissa of the i-th frame input image sequence;Y is the ordinate of the i-th frame input image sequence;Ii(x, y) is institute
State the gray value of the i-th frame input image sequence;IminFor the minimum value of the gray value of pixel in the i-th frame input image sequence;
ImaxFor the maximum value of the gray value of pixel in the i-th frame input image sequence;Step is drawing coefficient, 0 < step < 255.
4. the calculation method of intelligent terminal user visual perception response time delay index according to claim 1, feature exist
In carrying out target detection in the input image sequence, specifically include:
Mechanical arm target detection is carried out to the input image sequence using template matching method, acquisition detects the multiple of target
Input image sequence;Or
Finger target detection is carried out to the input image sequence using Otsu threshold method, acquisition detects the multiple defeated of target
Enter image sequence.
5. the calculation method of intelligent terminal user visual perception response time delay index according to claim 1, feature exist
Include: in, the step of determining first adaptive threshold and the second adaptive threshold
According to the i-th frame input image sequence and i+1 frame input image sequence calculating difference image sequence, and further calculate institute
The variance of error image sequence is stated, variance sequence is generated;
The variance of one group of preset quantity is chosen from the stem of the variance sequence, forms first group of variance, and from the variance sequence
The variance of preset quantity described in one group of the tail portion selection of column, forms second group of variance;
Maximum value in first group of variance is set as the first sub- adaptive threshold, by the minimum in first group of variance
Value is set as the second sub- adaptive threshold, and the described first sub- adaptive threshold and the second sub- adaptive threshold are determined as described
One adaptive threshold;
Maximum value in second group of variance is set as the sub- adaptive threshold of third, by the minimum in second group of variance
Value is set as the 4th sub- adaptive threshold, and the sub- adaptive threshold of the third and the 4th sub- adaptive threshold are determined as described
Two adaptive thresholds.
6. the calculation method of intelligent terminal user visual perception response time delay index according to claim 5, feature exist
In filtering out the first subsequence of multiple groups from the input image sequence according to the first adaptive threshold, and to the multiple groups first
Subsequence is ranked up, and to determine third picture numbers, specifically includes:
The 4th sub- adaptive threshold will be greater than in the variance sequence and less than the multiple groups of the sub- adaptive threshold of the third
The serial number of variance forms first position array positionend;
To the first position array positionendDiff is carried out, the first difference result Diff is generatedend(j), wherein
Diffend(j)=positionend(j+1)-positionend(j), j is the position of the variance sequence;
By the first difference result Diffend(j) brush is charged to after adding 1 in the corresponding position of value in greater than one second preset threshold
Screen judges array Pos;
When the brush screen judges array Pos for sky, by the first difference result Diffend(j) it is less than third predetermined threshold value in
Serial number charge to the first tail portion array pend;
By the first tail portion array pendIn minimum value corresponding to the serial number of input picture be determined as the third image
Serial number.
7. the calculation method of intelligent terminal user visual perception response time delay index according to claim 6, feature exist
In obtaining the brush screen and judge the first maximum value in array Pos when it is sky that the brush screen, which judges array Pos not,
frame3start;
By the first difference result DiffendIn be less than the third predetermined threshold value serial number charge to the second tail portion array p 'end;
By the second tail portion array p 'endIn be greater than or equal to the first maximum value frame3startThe serial number of corresponding serial number
Charge to third tail portion array pend1;
By the third tail portion array pend1In minimum value be determined as the third picture numbers.
8. the calculation method of intelligent terminal user visual perception response time delay index according to claim 6 or 7, feature
It is, is calculated by the following formula the error image sequence:
Di(x, y)=Ii+1(x, y)-Ii(x, y),
Wherein, x is the abscissa of the input image sequence;Y is the ordinate of the input image sequence;Ii(x, y) is described
The gray value of i-th frame input image sequence;Ii+1(x, y) is the gray value of the i+1 frame input image sequence;And
It is calculated by the following formula the variance of the error image sequence:
Wherein, n=h*w, h and w are respectively the height and width of the input image sequence,
9. the calculation method of intelligent terminal user visual perception response time delay index according to claim 6 or 7, feature
It is, the second subsequence of multiple groups is filtered out from the input image sequence according to the second adaptive threshold, and according to described first
Picture numbers, third picture numbers determine the second picture numbers from second subsequence of multiple groups, specifically include:
Median filtering is carried out to the variance sequence by an adaptive template, generates filtering rear difference sequence, it is described adaptive
The size of template are as follows: template=(frame3-frame1) %10, wherein frame1For the first image serial number,
frame3For the third picture numbers;
It obtains and is greater than the described second sub- adaptive threshold in the difference sequence of the filtering rear and is less than the described first sub- adaptive thresholding
The filtering variance serial number of the multiple groups variance of value, and further will in the filtering variance serial number be greater than the first image serial number and
Serial number less than the third picture numbers forms second position array positionstart;
To the second position array positionstartDiff is carried out, the second difference result Diff is generatedstart(k),
In, Diffstart(k)=positionstart(k+1)-positionstart(k), k is the position of the variance sequence;
By the second difference result Diffstart(k) serial number in less than the third predetermined threshold value charges to stem array pstart;
By the stem array pstartIn maximum value corresponding to the serial number of input picture be determined as the second image sequence
Number.
10. the calculation method of intelligent terminal user visual perception response time delay index according to claim 9, feature exist
In specifically including according to the first image serial number and the second picture numbers computing terminal response time:
When the terminal response time, is corresponding with the first image serial number equal to the second picture numbers corresponding time
Between difference.
11. the calculation method of intelligent terminal user visual perception response time delay index according to claim 10, feature
It is, calculates the application response time according to the third picture numbers and the second picture numbers, specifically include:
It is corresponding with second picture numbers that the application response time is equal to the third picture numbers corresponding time
Time difference.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610223448.9A CN105913429B (en) | 2016-04-12 | 2016-04-12 | The calculation method of intelligent terminal user visual perception response time delay index |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610223448.9A CN105913429B (en) | 2016-04-12 | 2016-04-12 | The calculation method of intelligent terminal user visual perception response time delay index |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105913429A CN105913429A (en) | 2016-08-31 |
CN105913429B true CN105913429B (en) | 2019-03-08 |
Family
ID=56745817
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610223448.9A Active CN105913429B (en) | 2016-04-12 | 2016-04-12 | The calculation method of intelligent terminal user visual perception response time delay index |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105913429B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107449977A (en) * | 2016-12-29 | 2017-12-08 | 福建奥通迈胜电力科技有限公司 | A kind of split-second precision difference computational methods based on image comparison technology |
CN107329883A (en) * | 2017-06-19 | 2017-11-07 | 中国信息通信研究院 | The automatic calculating method and system of intelligent terminal application program interaction response time delay |
CN111179408B (en) * | 2018-11-12 | 2024-04-12 | 北京物语科技有限公司 | Three-dimensional modeling method and equipment |
CN111858318B (en) * | 2020-06-30 | 2024-04-02 | 北京百度网讯科技有限公司 | Response time testing method, device, equipment and computer storage medium |
CN113334392B (en) * | 2021-08-06 | 2021-11-09 | 成都博恩思医学机器人有限公司 | Mechanical arm anti-collision method and device, robot and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101668223A (en) * | 2009-09-07 | 2010-03-10 | 航天恒星科技有限公司 | Method for measuring image transmission time delay |
CN103684889A (en) * | 2012-08-29 | 2014-03-26 | 云联(北京)信息技术有限公司 | Cloud computing-based speed measurement method applied to user terminal |
CN104679512A (en) * | 2015-02-12 | 2015-06-03 | 腾讯科技(深圳)有限公司 | Method and device for acquiring window program response time |
CN104965773A (en) * | 2015-07-09 | 2015-10-07 | 网易(杭州)网络有限公司 | Terminal, jamming detection method, device as well as game jamming detection method and device |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8387099B2 (en) * | 2002-12-10 | 2013-02-26 | Ol2, Inc. | System for acceleration of web page delivery |
-
2016
- 2016-04-12 CN CN201610223448.9A patent/CN105913429B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101668223A (en) * | 2009-09-07 | 2010-03-10 | 航天恒星科技有限公司 | Method for measuring image transmission time delay |
CN103684889A (en) * | 2012-08-29 | 2014-03-26 | 云联(北京)信息技术有限公司 | Cloud computing-based speed measurement method applied to user terminal |
CN104679512A (en) * | 2015-02-12 | 2015-06-03 | 腾讯科技(深圳)有限公司 | Method and device for acquiring window program response time |
CN104965773A (en) * | 2015-07-09 | 2015-10-07 | 网易(杭州)网络有限公司 | Terminal, jamming detection method, device as well as game jamming detection method and device |
Also Published As
Publication number | Publication date |
---|---|
CN105913429A (en) | 2016-08-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105913429B (en) | The calculation method of intelligent terminal user visual perception response time delay index | |
CN110047095B (en) | Tracking method and device based on target detection and terminal equipment | |
CN106846362A (en) | A kind of target detection tracking method and device | |
CN108229262B (en) | Pornographic video detection method and device | |
CN106920245B (en) | Boundary detection method and device | |
CN106127775A (en) | Measurement for Digital Image Definition and device | |
CN110674680B (en) | Living body identification method, living body identification device and storage medium | |
CN104978578A (en) | Mobile phone photo taking text image quality evaluation method | |
WO2019014813A1 (en) | Method and apparatus for quantitatively detecting skin type parameter of human face, and intelligent terminal | |
CN106650670A (en) | Method and device for detection of living body face video | |
CN110348385B (en) | Living body face recognition method and device | |
CN112307984B (en) | Safety helmet detection method and device based on neural network | |
CN111814846B (en) | Training method and recognition method of attribute recognition model and related equipment | |
CN111860568B (en) | Method and device for balanced distribution of data samples and storage medium | |
CN111191671A (en) | Electrical appliance waveform detection method and system, electronic equipment and storage medium | |
CN110111311B (en) | Image quality evaluation method and device | |
CN111881706B (en) | Living body detection, image classification and model training method, device, equipment and medium | |
CN110969045A (en) | Behavior detection method and device, electronic equipment and storage medium | |
WO2021068589A1 (en) | Method and apparatus for determining object and key points thereof in image | |
CN116912911A (en) | Satisfaction data screening method and device, electronic equipment and storage medium | |
CN111353526A (en) | Image matching method and device and related equipment | |
CN113139564A (en) | Method and device for training key point detection model, electronic equipment and storage medium | |
CN116977256A (en) | Training method, device, equipment and storage medium for defect detection model | |
CN115830002A (en) | Infrared image quality evaluation method and device | |
CN115471671A (en) | Network model training method, target recognition method and related equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20211230 Address after: 100191 No. 40, Haidian District, Beijing, Xueyuan Road Patentee after: CHINA ACADEMY OF INFORMATION AND COMMUNICATIONS Address before: 100191 No. 52 Garden North Road, Beijing, Haidian District Patentee before: CHINA ACADEME OF TELECOMMUNICATION RESEARCH OF MIIT |
|
TR01 | Transfer of patent right |